While for most people gambling is just an enjoyable experience without major consequences, for approximately 5% of adults, gambling can result in significant problems such as family, interpersonal and financial problems. Existing evidence, including work supported by this application's parent grant (R01HD060072, PI: Martins), demonstrates that PG is multifactorial and that it is influenced by a broad range of variables. However, PG research has been seriously constrained by the absence of longitudinal data and of a theory-driven and empirically-supported etiological model that integrates an array of risk and associated factors. This study integrates a developmental model of PG, Blaszczynski and Nower's Pathways model (so far only examined cross-sectionally) with a context -and intervention-informed developmental model of other risk behaviors (Ecodevelopmental theory). We propose to collect new information about gambling behaviors and PG in two time-points on a high-risk population of 1,100 youth/young adults from two population-based samples of Puerto Rican (ages 17-25) from two sites: the South Bronx (SBx), NYC and San Juan, Puerto Rico (PR) by taking advantage of an ongoing study, the Boricua Youth Study (BYS-R01DA033172, PI: Duarte and R01MH09837, PIs: Alegria, Canino & Duarte), which has been following Puerto Rican individuals prospectively since 2001. The specific aims of this study are to 1) Estimate: (1a) the prevalence (past year and past three- month) of any gambling, frequent gambling, and severity of PG during late adolescence/early adulthood; (1b) changes in the patterns of gambling and severity of PG over a two-year period when gambling behaviors are known to be undergoing the most changes; (1c) the characteristics of individuals who gamble (age, gender, age of onset of gambling and PG, and family history of PG), determining if PG estimates and characteristics differ between PR and SBx. 2) Examine developmental PG pathways from childhood to young adulthood, specifically (a) the relationship of externalizing, internalizing disorders and related risk factors with PG prevalence and change in late adolescence/young adulthood; (b) the influence of contextual factors (e.g. social support, neighborhood disadvantage) on PG pathways. 3) Understand how PG is related to other risk behaviors also being shaped during late adolescence/young adulthood, by determining different risk behavior clusters involving PG, HIV/STD risk behaviors, school drop-out, and antisocial behaviors. This innovative approach, using state of the art latent variable modeling, focuses on policy-modifiable risk factors within broader risk/resource contexts. By focusing on Puerto Ricans living in two settings we can also examine the important role of environment in PG in an ethnically homogeneous group. The proposed study, by building on existing and ongoing efforts, is a highly cost-effective way of generating unprecedented information about gambling behaviors among Puerto Ricans as well as examining the Pathways model longitudinally.